Dust collection..."Y" vs. "T"

[Jerry Lawrence]

First of all, let me apologize if this topic has been covered before (as it probably has).
My question involves dust collection and the use of a long straight run, interrupted by intersections and blast gates at each machine. Instead of using a "T" to branch off to each machine, would a "Y" connector restrict the flow less? The "Y" being straight on one side with the other branching off at around a 45 degree angle or a bit less. Just thinking that the "Y" would provide a less abrupt change in air flow, therefor providing for less loss in suction...or would it be enough to notice?
Thanks for any advice!

[Peter Gregory]

Yes, a Y allows much more airflow then a T. You are exactly right. I haven't designed a system for 5 years, so can't pull the % off the top of my head.

[Ole Anderson]

Wyes are the standard method of introducing another duct to your main run. In a suction based system, keeping your pipe losses as low as possible is the goal, without reducing the air velocity so much that dust will start collecting in the pipes.

[glenn bradley]

I think you have gleaned this from the responses but, never use a T; or a tight elbow for that matter. Smooth as possible sweeping flow path. Every abrupt change fouls the flow.

[Robert Payne]

Jerry,

Using wyes for each drop from your main duct is the more efficient method for airflow and it is also them possible to use a 45* elbow for your drop to your machine's blast gate. Remember to keep the blast gate fairly close to the machine and use a minimal amount of flexible duct to make the final connection to the dust port on the machine (many folks try to use solid duct all the way as long as the run is essentially straight pipe. My preference is 6" D-2729 PVC (sewer & drain) duct for the ducting for my 5 HP cyclone and 6" shop-made ports for major dust and chip producers (table saw, jointer, planer).

[Jerry Lawrence]

Thanks for confirming my assumption of Ys being more practical. I figured that was the case, but my local store carries Ts and not Ys, so that made me doubt myself a little. Now it just makes me doubt my local store.
Also, I have seen many people recommend (as Robert above) putting the blast gates close to the machine they collect dust from. I would think that putting the gate closer to the Y (assuming the trunk line was within easy reach) would reduce the distance of unnecessary air travel if a gate were closed. For instance, let's say I want to use the third machine on the line. The distance between the Y and each of the 'off' machines is five feet, making ten feet. If the gates were closed at the machines, the air has to fill (or maybe evacuate from) that extra ten feet on its way to the third machine. If the gates were right at the Y, the air would have less resistance as it wouldn't have that much extra space. Does any of that make sense, or am I just overthinking things that won't matter in real world application? Sometimes I think about things so much that I never actually get the job done!

[David Kumm]

Now that you get the importance of wyes, duct size is equally important to get right. If you run a 5 hp 15-16" impeller 6" mains are a little like putting in a T rather than a Y. A 3 hp 14" impeller goes with 6", 7-8" with the larger, at least for part of the run. Dave

[Michael W. Clark]

FYI,
If you are running 4000 FPM, then you would see a SP loss of 0.28"wg for the 45 branch fitting and 1.0"wg for the tee fitting.

If you are a single user (one branch per machine and one machine at a time) then the tee would be treated like a mitered 90, which would have 1.2"wg loss at 4000 FPM, when that branch is in operation.

The loss will change with the square of the velocity change. If you double the velocity, the loss will quadrouple, hence Dave's comments above.

I would put the gate in the branch where ever it is most convenient to operate. There will not be any additional resistance when the gate is closed because there will be no flow. The SP on the downstream side of the gate will equal the SP in the main where the branch ties in once the DC starts and reaches full speed.

Mike